Research map / Optimization / Distributed and compressed
Gradient Sparsification for Communication-Efficient Distributed Optimization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use magnitude-proportional gradient sparsification instead of uniform coordinate sampling at matched sparsity | Distributed SGD · Synthetic ℓ2-regularized logistic regression | objective suboptimality versus data passes | Helped |
| Use magnitude-proportional gradient sparsification instead of uniform coordinate sampling at matched sparsity | Distributed SVRG · Synthetic ℓ2-regularized logistic regression | objective suboptimality versus data passes | Helped |
| Use gradient sparsification instead of QSGD | Distributed SGD · Synthetic ℓ2-regularized logistic regression | objective suboptimality versus communication cost | Mixed |
| Sparsify gradients with κ = 0.004 | ADAM with CNN · CIFAR-10 | objective function versus communication cost | Helped |
| Use gradient sparsification in place of dense updates | Asynchronous multi-thread SGD · Synthetic SVM | training speed | Helped |
Closest papers by meaning
- The Convergence of Sparsified Gradient Methods (NeurIPS 2018 · similarity 0.91)
- Communication-efficient Distributed Learning for Large Batch Optimization (ICML 2022 · similarity 0.90)
- Efficient Distributed Learning with Sparsity (ICML 2017 · similarity 0.90)
- Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers (NeurIPS 2025 · similarity 0.89)
- Communication-Efficient Gradient Descent-Accent Methods for Distributed Variational Inequalities: Unified Analysis and Local Updates (ICLR 2024 · similarity 0.88)
- Towards Faster Decentralized Stochastic Optimization with Communication Compression (ICLR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.